University of Wollongong Research Online

Faculty of Science - Papers (Archive) Faculty of Science, Medicine and Health

2012

Behavioural adaptation of a from transient wetland specialist to an urban resident

John Martin University of Wollongong, [email protected]

Kristine O. French University of Wollongong, [email protected]

Richard Major Australian Museum

Follow this and additional works at: https://ro.uow.edu.au/scipapers

Part of the Life Sciences Commons, Physical Sciences and Mathematics Commons, and the Social and Behavioral Sciences Commons

Recommended Citation Martin, John; French, Kristine O.; and Major, Richard: Behavioural adaptation of a bird from transient wetland specialist to an urban resident 2012. https://ro.uow.edu.au/scipapers/4719

Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library: [email protected] Behavioural adaptation of a bird from transient wetland specialist to an urban resident

Abstract Dramatic population increases of the native white in urban areas have resulted in their classification as a nuisance . In response to community and industry complaints, land managers have attempted to deter the growing population by destroying ibis nests and over the last twenty years. However, our understanding of ibis ecology is poor and a question of particular importance for management is whether ibis show sufficient site fidelityo t justify site-level management of nuisance populations. Ibis in non-urban areas have been observed to be highly transient and capable of moving hundreds of kilometres. In urban areas the population has been observed to vary seasonally, but at some sites ibis are always observed and are thought to be behaving as residents. To measure the level of site fidelity, we colour banded 93 adult ibis at an urban park and conducted 3-day surveys each fortnight over one year, then each quarter over four years. From the quarterly data, the first eary resighting rate was 89% for females (n = 59) and 76% for males (n = 34); this decreased to 41% of females and 21% of males in the fourth year. Ibis are known to be highly mobile, and 70% of females and 77% of males were observed at additional sites within the surrounding region (up to 50 km distant). Our results indicate that a large proportion of ibis have chosen residency over transience both within the study site and across the broader urban region. Consequently the establishment of refuge breeding habitat should be a priority localised management may be effective at particular sites, but it is likely to have an impact across the broader population.

Keywords urban, resident, adaptation, behavioural, bird, transient, wetland, specialist

Disciplines Life Sciences | Physical Sciences and Mathematics | Social and Behavioral Sciences

Publication Details Martin, J., French, K. O. & Major, R. (2012). Behavioural adaptation of a bird from transient wetland specialist to an urban resident. PLoS ONE, 7 (11),

This journal article is available at Research Online: https://ro.uow.edu.au/scipapers/4719 Behavioural Adaptation of a Bird from Transient Wetland Specialist to an Urban Resident

John Martin1,2*, Kris French1, Richard Major2 1 Institute for Conservation Biology and Environmental Management, School of Biological Sciences, University of Wollongong, Wollongong, New South Wales, Australia, 2 Terrestrial Ecology, Australian Museum, Sydney, New South Wales, Australia

Abstract Dramatic population increases of the native white ibis in urban areas have resulted in their classification as a nuisance species. In response to community and industry complaints, land managers have attempted to deter the growing population by destroying ibis nests and eggs over the last twenty years. However, our understanding of ibis ecology is poor and a question of particular importance for management is whether ibis show sufficient site fidelity to justify site-level management of nuisance populations. Ibis in non-urban areas have been observed to be highly transient and capable of moving hundreds of kilometres. In urban areas the population has been observed to vary seasonally, but at some sites ibis are always observed and are thought to be behaving as residents. To measure the level of site fidelity, we colour banded 93 adult ibis at an urban park and conducted 3-day surveys each fortnight over one year, then each quarter over four years. From the quarterly data, the first year resighting rate was 89% for females (n = 59) and 76% for males (n = 34); this decreased to 41% of females and 21% of males in the fourth year. Ibis are known to be highly mobile, and 70% of females and 77% of males were observed at additional sites within the surrounding region (up to 50 km distant). Our results indicate that a large proportion of ibis have chosen residency over transience both within the study site and across the broader urban region. Consequently the establishment of refuge breeding habitat should be a priority localised management may be effective at particular sites, but it is likely to have an impact across the broader population.

Citation: Martin J, French K, Major R (2012) Behavioural Adaptation of a Bird from Transient Wetland Specialist to an Urban Resident. PLoS ONE 7(11): e50006. doi:10.1371/journal.pone.0050006 Editor: Frederick R. Adler, University of Utah, United States of America Received May 29, 2012; Accepted October 18, 2012; Published November 21, 2012 Copyright: ß 2012 Martin et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: We thank the W.V. Scott Charitable Trust, Waste Services New South Wales (NSW) Environmental Solutions, the Centennial Parklands Foundation, the NSW Office of Environment and Heritage and the Australian Museum for financial support. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected]

Introduction of Victoria between 1955 and 1980 identified 63 sites and concluded that site use, the number of and the timing of Understanding site fidelity is important for the conservation of breeding were annually variable [15]. Breeding occurred in an rare species [1] and the management of introduced pest or average of 7.660.9 se of 25 years per site, with the amount of nuisance native species [2,3]. Site fidelity is likely to be affected by breeding varying from tens of birds up to ,20 000 based on the resource availability, which influences a species’ movements and data provided for 25 colonies [15]. This annual variability reflects behaviour [4]. The urban environment appears to have abundant the unpredictable Australian climate where there is strong habitat and foraging resources for a range of species and these selection for native species to evolve behavioural plasticity resources are being utilised by both rare [5,6], and pest species [16,17]. Consequently ibis have been observed to breed sporad- [7,8]. Behavioural flexibility to novel food resources is a mecha- ically at particular sites and at any time of the year as conditions nism that assists species to colonise new habitats [9]. As a result of allow [14,15]. Long distance movements between unpredictable high resource availability, individuals adapting to urban habitats habitat is likely to be energetically costly and may have an impact are predicted to alter their behaviour, influencing rates of on survival and fecundity [18], thus selection should favour site immigration, emigration, survival and fecundity [10]. High fidelity over transience where resources are consistently abundant resource levels in cities relative to natural habitats may cause [19,20], such as in urban areas. Increased site fidelity has been some species to increase their population sizes. found in urban habitats for some synanthropic species [7,21,22], The ( molucca), hereafter referred but little information is available on site fidelity within urban ibis to as ‘‘ibis’’, is considered a nuisance native species in urban areas populations. because its population has increased and encroaches upon humans Site fidelity has been defined as the degree to which an (see ‘Study species and site’) [11,12]. In response, land managers returns to a specific site. Studies commonly assess site fidelity with have destroyed ibis nests and eggs to deter breeding and reduce respect to returning to breeding grounds [19,23] or recruitment [13]. Traditionally, large numbers of ibis were moving through migratory staging grounds [24,25]. Ecologically, observed breeding within flooded inland wetlands, and movements studies assessing site fidelity are questioning behavioural decision- of hundreds of kilometres were observed between these ephemeral making with respect to patch quality [18,24,26], and this question resources [14,15]. An assessment of ibis breeding across the State is being applied to resource availability in urban habitats [27,28].

PLOS ONE | www.plosone.org 1 November 2012 | Volume 7 | Issue 11 | e50006 Behavioural Adaptation of an Urban Bird

Understanding the site fidelity of ibis is important to inform that humans are at risk of infections from birds living in close both management actions aimed at resolving human-wildlife proximity to humans [37]. Thirdly, people object to the smell, conflicts, and for conservation. The urban environment is thought noise, defecation and scavenging of ibis in urban environments. to be providing a refuge for ibis [8] as the inland wetland habitat Finally, the ingestion of an ibis by in a jet engine in 1995 led to the has been degraded from human modification of river flows realization that ibis colonies near airports presented a major [29,30]. Ibis are long-lived (up to 26 years, [31]) and the breeding aviation hazard. During the late 1990’s the ibis population within success of urban populations may be important for maintaining the study site was reduced from a seasonal peak of over 1000 down populations of this species at a continental scale, as only a few to 400 ibis through the removal of nesting habitat and the small natural breeding events have been documented over the last destruction of nests [J. Cartmill pers. comm.]. No habitat or nest decade [32,33]. In this study we assess the site fidelity of ibis to an destruction has been implemented at the study site since 2005. urban park, defining high site fidelity as presence on greater than However, the management of nesting ibis within urban parks, or equal to 75% of surveys within a calendar year. industrial land and residences is commonplace across the Sydney As abundant foraging and habitat resources are consistently- region to mitigate local complaints. Foraging by ibis within present within the urban environment [7], in contrast to the landfills is thought to be sustaining the large urban population and ephemeral inland wetlands [16,17], a high degree of site fidelity the extended breeding season [12], which cause the major issues might be expected in ibis that occupy cities. The aims of this study for the general public. are 1) to determine levels of site fidelity of male and female ibis, 2) to assess the extent of movements to foraging sites and breeding Field Surveys colonies within the study region, and 3) to measure fluctuations in Adult ibis were individually caught (n = 34 males and 59 the population at the study site over a 4-year period. females) within the park between July 2005 and April 2006 using a hand-held foot noose made of fishing line with bread as an Methods attractant. Ibis were individually colour-banded with a combina- tion of four bands on the tarsus and tibia [38]. The sex of each Ethics Statement bird was determined based upon bill length (exposed culmen, Bird capture, banding and surveys were conducted in accor- [34]). dance with the University of Wollongong Animal Care and Ethics Surveys of the ibis population within the park were conducted Committee (AE05-15r8); Australian Museum Animal Care and over three consecutive days. During these surveys we aimed to re- Ethics Committee (08-05); New South Wales National Parks and sight banded ibis and count the total number of adults and Wildlife Service Scientific Licences (S12568, S12586) and an juveniles in the park. A single observer traversed a route that Australian Bird and Bat Banding Scheme research project (1209). covered all areas where ibis occurred (all but two of the surveys were conducted by the same observer). Surveys were conducted Study Species and Site for three hours during the afternoon (,3pm to ,6pm, depending Ibis are large wetland birds with mostly white , on day-length) when ibis were most abundant [39] using a wingspan to 125 cm, a long down-curved bill (to 22 cm) and binoculars and a spotting scope. Between January 2006 and April a black unfeathered head and neck (only in adults $3years) [34]. 2007 fortnightly 3-day surveys (n = 32) were conducted to in- Up to 7000 adult ibis are known to occur within the Sydney tensively assess site fidelity. Following this, 3-day surveys were region, Australia, throughout the year [12], and they are conducted quarterly (Jan, Apr, July, Oct) until the end of 2009 frequently observed in city parks. Centennial Park (33u5399569S, (n = 16, including four from 2006). From March 2007 we also 151u1499019E), near the centre of Sydney, is a large urban park counted the number of active nests (defined as containing eggs, (186 hectare) containing ponds of varying size (26 ha total area), nestlings or under incubation) and from September 2007 we noted some of which contain small islands. Ibis are always present and if the legs, and therefore colour bands, of adult ibis were nest on the islands and forage within the grass and pond habitats. unidentifiable due to obstructions (e.g. water, vegetation or nest In addition, people commonly feed the birds (e.g. bread) within the material). Banded ibis were also resighted outside of the study site park, and scavenging occurs from bins and picnickers. during monthly surveys of the Sydney region. These surveys were Ibis have the capacity to be highly mobile. During daily conducted over three consecutive days by a single observer and foraging, multiple movements of 109s of kilometers between included the largest foraging and breeding locations as well as roosting and foraging resources have been recorded during radio smaller sites (n = 54) where ibis occur in the Sydney region. At all telemetry [8]. Larger movements of 100’ s of kilometres have been sites, we recorded the number of adults, juveniles, nestlings and inferred through the colonisation of distant breeding sites following nests (for further detail see [12]). the flooding of a wetland [14]. At the site scale the disappearance of a bird could be the result of a movement within the surrounding Statistical Analysis region (10’ s km) or outside the region (100 km’s). For ibis The survival rate of colour-banded ibis and rates of transient movements are likely to involve leaving an area for an temporary emigration and immigration were estimated for both extended period (e.g. months) and population surveys of the sexes using the robust design [40] within Program MARK [41]. Sydney region have identified that adult ibis seasonally immigrate By incorporating ‘replicate’ surveys in the same survey period to the region and disperse following breeding [12]. This seasonal (i.e. surveys on three consecutive days) the robust design is able population increase, which is associated with breeding, fuels to produce estimates of the temporary emigration and community complaints. immigration rates. The parameters of emigration and immigra- Ibis are considered to present a pest problem for four main tion allow for, and estimate, the likelihood of marked individuals reasons. Firstly, their population increase in urban areas over the moving in and out of the study location between survey periods last 30 years [12] has lead to the misconception that they are the (e.g. quarterly surveys). This in turn improves the survival African sacred ibis (T. aethiopicus) and should be managed as a non- estimation and our interpretation of site fidelity. The robust native pest (as is the case in Europe and North America [35,36]). design estimates five parameters in each model (Table 1 & 2): Secondly, ibis are known to carry diseases and there is a concern survival rate (Q), emigration rate (c99), immigration rate (c9),

PLOS ONE | www.plosone.org 2 November 2012 | Volume 7 | Issue 11 | e50006 Behavioural Adaptation of an Urban Bird initial capture probability (p) and recapture probability (c). Each Results of these parameters was estimated in a range of models that included the effects of sex (g) and time (t). A period (.) Short-term Survival and Residency represents a variable that is held constant e.g. the exclusion of The best fitting model for the 15-months of fortnightly surveys time-specific effects. Model selection was based on Akaike’s included a constant time parameter for survival, emigration, Information Criterion [42]. immigration and the capture and recapture probabilities. This Studies reporting site fidelity commonly provide a species- model was superior to models that included separate parameters specific definition [18,43]. For this study we have defined high site for each time interval (Table 1). Using this model, the survival rate fidelity as birds observed on greater than or equal to 75% of the for both females and males was estimated to be close to 100% fortnightly (conducted over 15-months, n = 32) and quarterly (female = 0.9960.005 se; male = 0.9860.013) over the 15-months surveys (n = 4 within a calendar year). Site fidelity was measured of fortnightly surveys. Because ibis are thought to be long-lived for each sex separately, by estimating the frequency with which a high survival rate was expected. Of greater interest is the male and female ibis were resighted within the study site. The estimated probability of emigration, that is, an observed banded percentage of surveys in which an individual was resighted was ibis not being present during the next survey. The model estimated amalgamated into the percentile categories of 0, 1 to 25, 26 to 50, that 17% (60.021) of females and 21% (60.031) of males would 51 to 75 and 76 to 100, for both the fortnightly and quarterly data. leave the study site between fortnightly surveys. Of the individuals For the fortnightly surveys (n = 32) we calculated the frequency that were not observed during a survey, the model estimated that that males and females were resighted and measured comparative 86% (60.022) of females and 94% (60.020) of males would not be site fidelity with respect to sex using Chi-square analysis. We then resighted during the next survey. This means that unobserved ibis looked at both the breeding (July to December) and the non- are unlikely to be resighted within the study site for an extended breeding (January to June) seasons [12,39,44] to determine period. whether sex specific differences in residency varied seasonally. Corroborating the MARK analysis, the female resighting 2 For the quarterly surveys we calculated the frequency with which frequency was greater than that of males (x4 = 15.8, p = 0.003) both male and female ibis were resighted in each year (2006 to indicating greater site fidelity over the 15-month fortnightly survey 2009) and determined sex specific differences in the annual site period (Fig. 1). A high level of site fidelity (i.e. resighted in greater fidelity, again using Chi-square analysis. We also used Chi-square than or equal to 75% of surveys) was observed for a small number analyses to determine whether male or female ibis were more likely of females (9% n = 59) and males (6% n = 34). The site fidelity of to be resighted outside of the study area and whether they differed female ibis appeared to be greater than that of males during the in their use of the major habitat categories of landfills, parks or non-breeding season (January to June) with 75% (n = 59) of females being resighted compared to 59% (n = 34) of males, but breeding colonies. 2 The relationships between ibis abundance within the study site, this difference was not statistically significant (x4 = 8.02, p = 0.09). the season (non-breeding, breeding) and the year (2006, 2007, Similarly, the site fidelity of females appeared to be greater than 2008 and 2009) were assessed using a two-factor analysis of that of males during the breeding season (July to December) with variance (ANOVA). Separate analyses were conducted for adults 77% (n = 59) of females being resighted compared to 47% (n = 34) of males, but again, the difference was not statistically significant and juveniles. The breeding season of adult ibis was defined as July 2 to December based on nesting records, but breeding has also been (x4 = 9.21, p = 0.056). observed at low levels all year round [12,44]. The analysis of juveniles was lagged by one month to account for the delay Long-term Survival and Residency between laying and fledging. Annual variation in reproductive The best fitting model for the 4-year quarterly surveys included activity, as measured by counting the number of nests at the study a constant time parameter for survival, emigration, immigration site during three breeding seasons (2007, 2008, 2009), was and the capture and recapture probabilities. This model was compared with a single-factor ANOVA. The relationship between superior to models that included separate parameters for each time the number of unidentifiable adult ibis and season (non-breeding, interval (Table 2). Over the 4-year study a survival rate of 92% breeding) was investigated using a single-factor ANOVA, as the (60.014) was estimated for female ibis and 87% (60.028) for resighting rate of banded birds may be reduced due to nest males by the best fitting model. Between quarterly surveys, the incubation (these data are a subset of the total population count model estimated that 33% (60.036) of females and 42% (60.088) assessed above). All data were tested for homogeneity of variance of males would emigrate from the study site. Of the individuals with Cochran’s test and, where necessary, log(x+1) transformations that were not observed during a survey, it was estimated that 66% were used to stabilise variances. Significant interactions were (60.048) of females and 74% (60.075) of males would not be identified with Tukey’s pairwise comparisons. resighted during the subsequent survey. The number of banded ibis that were resighted in each survey decreased over the four years and there was little evidence of

Table 1. Best-fitting models of resighted banded ibis assessed using the robust design within Program MARK.

Model* AICc DAICc Model Likelihood Num. Par Deviance

{Q(g,.), c99(g,.), c9(g,.), p(g,.), c(g,.)} 5048 0 0.99967 127 7143 {Q(g,t), c99(g,.), c9(g,.), p(g,.), c(g,.)} 5064 16 0.00029 151 7101 {Q(g,.), c99(g,.), c9(g,.), p(g,.), c(g,t)} 5099 51 0.00004 173 7080

Surveys were conducted (a) fortnightly (n = 32) over 15-months. doi:10.1371/journal.pone.0050006.t001

PLOS ONE | www.plosone.org 3 November 2012 | Volume 7 | Issue 11 | e50006 Behavioural Adaptation of an Urban Bird

Figure 1. Residency of male (black, n = 34) and female (white, n = 59) ibis over 15-months from pooled fortnightly 3-day surveys (n = 32). doi:10.1371/journal.pone.0050006.g001

a seasonal effect on sex-based differences in site fidelity (Fig. 2). increased over the three years (F2,15 = 6.39, P = 0.01) (Fig. 3). More The number of resighted female and male ibis was not significantly nesting occurred during the 2009 breeding season compared with 2 2 different during 2006 (x4 = 8.39, p = 0.078) and 2009 (x4 = 5.22, that of 2007 (Tukey pairwise, P = 0.009) and 2008 (Tukey p = 0.265); however, a significantly higher number of females were pairwise, P = 0.058); levels of nesting in the latter two years were 2 2 resighted in 2007 (x4 = 10.32, p = 0.035) and 2008 (x4 = 13.03, comparable (Tukey pairwise, P = 0.63). The number of juveniles p = 0.011). A greater percentage of females displayed high site differed between years (F3,31 = 4.18, P = 0.013, Fig. 3) with fidelity, defined as being observed in greater than or equal to 75% significant increases occurring during the study. Significantly of surveys in each year. Thirty-seven percent (n=59) of females higher numbers of juveniles were observed between August to and 18% (n=34) of males showed high site fidelity in 2006, 47% of January than February to July (F1,31 = 8.77, P = 0.006), and there females and 21% of males in 2007, 26% of females and 15% of was no significant interaction between season and year males in 2008 and 16% of females and 9% of males in 2009. (F3,31 = 0.34, P = 0.80). Throughout the study we were unable to determine whether every ibis was colour-banded due to occasional visual obstruction Resightings Beyond the Study Site of their legs. In 2008 and 2009 the number of unidentifiable adult Ibis banded in Centennial Park were commonly resighted in ibis within the population during a survey of the study site varied surveys across the Sydney region (for survey methods, see [12]), 2 between a minimum of 2% and a maximum of 38% (n = 3 and with both sexes equally likely to be resighted (x1 = 0.23, p = 0.63). n = 137 unidentifiable ibis, respectively). Unsurprisingly, the pro- We observed 70% (n=59) of colour-banded females at one or portion of adult ibis with unidentifiable legs was significantly more of 32 different sites away from the study site and 77% of greater during the breeding season (July to December) when some males (n=34) at one or more of 20 sites. Landfills were a major birds were sitting on nests (F1,22 = 36.27, P,0.000). foraging resource with comparable usage by birds of each sex 2 (x1 = 0.43, p = 0.51); we encountered 41% of females (n = 59) and Population Size 50% of males (n = 34) at least once at one or more of the five open The number of adult ibis using the study site did not vary landfills in the region. For example, 19% of females and 35% of significantly between years (F3,32 = 2.4, P = 0.08, Fig. 3), but was males were observed at least once at a landfill 20 km to the north, greater in the breeding seasons than in the non-breeding seasons and 24% of females and 18% of males were observed at least once (F1,32 = 6.55, P = 0.015). There was no significant interaction at a landfill 35 km to the west of the study site. Other locations at between season and year, with the same seasonal fluctuations which ibis were frequently seen in the Sydney region included detected in each of the four years (F3,32 = 1.9, P = 0.15). During the parkland and breeding colonies. Parks provided similar resources breeding seasons (July to December) the number of nests observed to the study site yet only 36% of females and 29% of males were

Table 2. Best-fitting models of resighted banded ibis assessed using the robust design within Program MARK.

Model* AICc DAICc Model Likelihood Num. Par Deviance

{Q(g,.), c99(g,.), c9(g,.), p(g,t) = c(g,t)} 2348 0 0.9943 98 2871 {Q(g,t), c99g,.), c9(g,.), p(g,.), c(g,.)} 2359 11 0.0050 67 2863 {Q(g,t), c99(g,t), c9(g,.), p(g,.), c(g,.)} 2363 15 0.0007 133 2783

*The robust design estimates five parameters in each model: survival rate (Q), emigration rate (c’’ ), immigration rate (c’ ), initial capture probability (p) and recapture probability (c). The symbols within parentheses indicate whether the parameter is held constant ‘.’ or whether separate parameters are include for each sex (g) or survey interval (t). Surveys were conducted quarterly (n = 16) over 4-years. doi:10.1371/journal.pone.0050006.t002

PLOS ONE | www.plosone.org 4 November 2012 | Volume 7 | Issue 11 | e50006 Behavioural Adaptation of an Urban Bird

Figure 2. Resighting rate of colour-banded male (solid, n = 34) and female (dash, n = 59) ibis over 4-years from pooled quarterly 3- day surveys. doi:10.1371/journal.pone.0050006.g002

2 observed therein (x1 = 0.82, p = 0.36). Interestingly, at breeding were not encountered. Secondly, observations from across the colonies we observed twice as many males (29%) as females (14%) Sydney region indicated that some ibis make foraging trips out of the 2 at least once, although this was not significant (x1 = 2.53, study site during the day, but return for some part of each day. These p = 0.11). birds may have been outside the study site at the time of the survey but still showed site fidelity at the temporal scale of one day [8]. Discussion Thirdly, birds may have taken up residence at another site, either temporarily or permanently. By conducting surveys on three For species able to exploit them, urban environments have the consecutive days (robust design, [41]) we were able to estimate the attributes of abundant habitat and consistent resources [22,45,46]. rates of temporary emigration and immigration, distinguishing birds The behaviour of species that have colonised urban environments in the third category from the other two sources of error. The model has often been observed to change from that of their ‘natural’ estimated that between each fortnightly set of surveys ,20% of behaviour [5–7,47]. For example, studies of European Blackbirds observed ibis would not be resighted in the next survey (temporary have found that urban birds are more sedentary than rural birds emigration). Similarly, between quarterly surveys ,33% of the [47,48]. Approximately one quarter of the banded ibis in this study observed ibis would not be resighted in the next survey. Movements were observed to exhibit a high level of site fidelity in three of the out of the study site may be daily foraging or transient movements for four years, and given our inability to detect all banded birds in extended periods (e.g. months), however our data suggest that birds a survey, this figure represents the minimum estimate. This level of are likely to return. Evidence for this is provided by the two site fidelity observed in this study represents a dramatic beha- immigration estimates. For the fortnightly survey ,90% of the vioural adaptation by ibis from a transient species moving between banded ibis that were absent from a survey were unlikely to be ephemeral wetlands (i.e. no site fidelity) to a resident within resighted in the next fortnightly survey, but for the quarterly surveys a relatively constant urban environment. This finding has this dropped to ,65%. This suggests that ibis that ended a period of important implications for the conservation and management of residency [43] were more likely to stay away from the site for a couple ibis. The high level of site fidelity suggests that birds from colonies of months than for a couple of weeks. occurring at locations distant from priority human assets (e.g. Female ibis showed greater site fidelity than males. A higher airport flight paths) are unlikely to have a major impact on people. proportion of females were resighted each year and they were These locations could effectively be managed as refuge habitat [8] resighted more frequently than males throughout each year. There and, with the establishment of designated refuges, the manage- was some indication that males visited external breeding colonies ment of priority nuisance colonies is less likely to have a serious more frequently than females, although this result was not impact on the regional productivity of urban ibis. significant. These movements throughout the study might be associated with a bird’s assessment of potential reproductive Levels of Site-fidelity opportunities. Ibis have been observed to engage in polygamy and Birds that were known to be alive but not seen on a particular extra-pair copulation, although breeding pairs that share parental survey could have been absent for three reasons. Firstly, some birds investment equally have been observed to fledge the most offspring would have been present in the study site, but unseen by the [34,44]. observer, either because their legs were not visible, or because they

PLOS ONE | www.plosone.org 5 November 2012 | Volume 7 | Issue 11 | e50006 Behavioural Adaptation of an Urban Bird

Figure 3. Ibis within Centennial Park showing the annual fluctuations associated with the breeding cycle for adults (long-dash), juveniles (solid) and nests (dash). doi:10.1371/journal.pone.0050006.g003

Use of Alternative Sites management and conservation. The ability of ibis to move long The habitat type where we detected most birds outside the study distances is well known and band recoveries have identified site was landfills. Landfills represent a major foraging resource for movements of 1000’s of kilometres by dispersing juveniles [14,52]. urban ibis [12] and we observed 41% of females and 50% of males However, we have a limited understanding of where adult ibis foraging within open landfills at least once. There is no doubt that disperse and less is known about the site fidelity or natal philopatry the percentage of ibis observed within landfills is an underestimate of the birds that do disperse. Of particular interest are the possible [8,38] as our surveys covered only a fraction of their operating interactions between the urban and the inland ‘natural’ popula- hours, and leg bands are difficult to detect in this circumstance. tions, but to date, no birds have been recorded transitioning Substantially higher resighting rates of marked birds have been between these populations. (The ‘natural’ inland habitats are recorded with patagial tags compared with leg bands [38,49], difficult to access and have a very low human population density; including within landfills. The abundant resources within landfills the ibis population estimates are derived from annual aerial are likely to be fueling increases in the urban ibis population, the surveys [16,32].) extension of the breeding season [12] and the resultant human- Ibis are believed to be long-lived (up to 26-years [31]), which is wildlife conflicts. Similar observations of the relationship between supported by the high rate of adult survival estimated over the four years of this study (females 92% and males 85%). Urban ibis population increases and conflicts have been made for other colonies may be subjected to management, primarily by suppres- nuisance species [45,50]. sing recruitment through nest and egg destruction [13]. The aim of this management is to reduce human-wildlife conflicts by Population Variability decreasing the number of nesting birds or deterring nesting The ibis population within the study site followed seasonal entirely. However, the continual suppression of recruitment in trends of growth and decline in each year of the study, a trend also urban areas may have long-term implications as the non-urban observed for the regional population [12]. In association with the ibis population has been declining over the last 20-years [32]. breeding season we observed an influx of adult ibis probably from Over the last decade there has been an absence of large ‘natural’ within the study region as well as beyond. Towards the end of the breeding events as a direct result of drought [53] and insufficient breeding season we observed the adult and juvenile population to environmental flows [33], although the urban population has decline significantly. Again some birds would disperse within the grown over the same period [12]. Importantly, there may be region while it is likely that others left the Sydney region. For a (presently unknown) time lag between when reproductive failure example, we have opportunistic observations of ibis (n = 17) and a population reduction is evident in the adult population. outside the study region at locations between 50 km and 400 km Therefore, breeding within the urban environment may be (mean = 84 km (623 se)) distant from the study site. Thus, the important for the persistence of this species. movement behaviour of individuals has implications for popula- Currently we lack knowledge of the recruitment rate of ibis tion dynamics at a broader scale [1,51] and therefore for fledglings to the reproductive population (third year) in both urban

PLOS ONE | www.plosone.org 6 November 2012 | Volume 7 | Issue 11 | e50006 Behavioural Adaptation of an Urban Bird and natural habitats. Ibis achieve good rates of nesting success in required to achieve appreciable population reduction at a partic- urban areas, with the percentage of eggs laid that fledged in two ular local site would have an unacceptable impact on the species, colonies ranging from 25% (n = 541) to 40% (n = 343, [52]), but given its current, urban-biased distribution [12,32]. The best comparable data are yet to be measured within the ephemeral option for managing nuisance colonies is to remove nesting habitat inland wetlands (non-urban). It is likely that the breeding success in (usually exotic palms) forcing the colony into alternative habitat. urban and ‘natural’ environments may differ [10,54]. The survival This approach, however, depends on the availability of suitable rate in the resource-rich urban environment might be expected to alternate habitat that can support a large colony and that is distant be higher than in natural systems, where variable foraging and the from sensitive areas (e.g. airports). Such habitat exists but there is distance between resources may negatively affect survival [54]. often resistance from nearby human residents to accept ibis Conversely, pollutants present in urban foraging sites may result in because they perceive ibis to be a global rather than local pest. If lower levels of reproductive success [55]. The non-urban rates of ibis are prevented from settling in large areas of suitable habitat, survival, fecundity and particularly colonisation by ‘urban’ ibis colonies are likely to fracture and nest in multiple locations near to could have important implications for the longevity of the non- those from which nesting habitat has been removed. Not only will urban ibis population. Thus recruitment and dispersal from the this fail to alleviate the risk to aircraft, but it is likely to create urban environment may be very important, as has been observed additional problems with humans. Identifying areas within a region for Scrub Jays and American Crows [10,46]. in which ibis will be free from persecution is therefore a fundamental component of ibis management. In the longer Effectiveness of Management term, reduction of food availability at landfills is likely to make Wildlife management for population [56] and disease control urban environments less attractive and may encourage ibis to settle [57,58] can result in unexpected and potentially undesirable in areas further from human habitation [8]. outcomes. Given that collision with ibis presents an unacceptable risk to jet aircraft, management of colonies or birds which traverse Acknowledgments aircraft approach paths is necessary to reduce the collision risk. Killing large numbers of adult birds as a solution may have an We thank Geoff Ross for field assistance and two anonymous reviewers for unacceptable effect on the broader ibis population given the level their constructive comments. of population connectivity [8] and the absence of recent breeding events in their traditional habitat [33]. Nest and egg destruction Author Contributions directed towards reducing population size is an inefficient method Conceived and designed the experiments: JM KF RM. Performed the of effecting local level population reduction because of the high experiments: JM RM. Analyzed the data: JM. Contributed reagents/ level of population interchange at a regional scale. The level of materials/analysis tools: JM KF RM. Wrote the paper: JM KF RM. nest and egg destruction at the regional scale that would be

References 1. Jorgensen SJ, Reeb CA, Chapple TK, Anderson S, Perle C, et al. (2010) 15. Cowling SJ, Lowe KW (1980) Studies of in Victoria, I: records of breeding Philopatry and migration of Pacific white sharks. Proc. R. Soc. B 277: 679–688. since 1955. Emu 81: 33–39. 2. Grund MD, McAninch JB, Wiggers EP (2002) Seasonal movements and habitat 16. Kingsford RT, Wong PS, Braithwaite LW, Maher MT (1999) Waterbird use of female white-tailed deer associated with an urban park. The Journal of abundance in eastern Australia, 1983–92. Wildlife Research 26: 351–366. Wildlife Management 66, 123–130. 17. Kingsford RT, Norman FI (2002) Australian waterbirds – products of the 3. Prange S, Gehrt SD, Wiggers EP (2003) Demographic factors contributing to continent’s ecology. Emu 102: 47–69. high raccoon densities in urban landscapes. Journal of Wildlife Management 67: 18. Hoover JP (2003) Decision rules for site fidelity in a migratory bird, the 324–333. prothonotary warbler. Ecology 84: 416–430. 4. Saggese K, Korner-Nievergelt F, Slagsvold T, Amrhein V (2011) Wild bird 19. Switzer PV (1993) Site fidelity in predictable and unpredictable habitats. Evol. feeding delays start of dawn singing in the great tit. Behavioral Ecology 22: 39– Ecol. 7: 533–555. 45. 20. Fuxjager MJ, Montgomery JL, Becker EA, Marler CA (2010) Deciding to win: 5. Howell RG (1982) The urban coyote problem in Los Angeles County. In: Marsh interactive effects of residency, resources and ‘boldness’ on contest outcome in RE, editor. Proceedings of the 10th Vertebrate Pest Conference. University of white-footed mice. Animal Behaviour 80: 921–927. California: Davis, CA. 21–23. 21. Johnston RF (2001) The synanthropic birds of North America. In: Marzluff JM, 6. van der Ree R, McDonnell MJ, Temby I, Nelson J, Whittingham E (2006) The Bowman R, Donnelly RE editors. Avian ecology and conservation in an establishment and dynamics of a recently established urban camp of flying foxes urbanizing world. Kluwer Academic, Norwell, MA. 49–67. (Pteropus poliocephalus) outside their geographic range. Journal of Zoology 268: 22. Withey JC, Marzluff JM (2009) Multi-scale use of lands providing anthropogenic 177–185. resources by American crows in an urbanizing landscape. Landscape Ecology 7. McKinney ML (2002) Urbanization, biodiversity, and conservation. BioScience 24: 281–293. 52: 883–890. 23. Cohen JB, Gratto-Trevor C (2011) Survival, site fidelity, and the population 8. Martin JM, French K, Ross GR, Major RE (2011) Foraging distances and dynamics of piping plovers in Saskatchewan. Journal of Field 82: habitat preferences of a recent urban coloniser: the Australian white ibis. 379–394. Landscape and Urban Planning 102: 65–72. 24. Hestbeck JB, Nichols JD, Malecki RA (1991) Estimates of movement and site 9. Sol D, Lefebvre L (2000) Behavioural flexibility predicts invasion success in birds fidelity using mark-resight data of wintering Canada geese. Ecology 72: 523– introduced to New Zealand. Oikos 90: 599–605. 533. 10. Bowman R, Marzluff JM (2001) Integrating avian ecology into emerging 25. Fox AD, Hilmarsson JO´ , Einarsson O´ , Walsh AJ, Boyd H, et al. (2002) Staging paradigms in urban ecology. In: Marzluff JM, Bowman R, Donnelly RE editors. site fidelity of Greenland white-fronted geese Anser albifrons flavirostris in Iceland: Avian ecology and conservation in an urbanizing world. Kluwer Academic, based upon resighting histories of marked individuals, a high level of site loyalty Norwell, MA. 569–578. was found for Greenland white-fronted geese staging in Icelandic stopover areas 11. Murray NJ, Shaw PP (2006) Breeding biology of the Australian white ibis in spring and autumn. Bird Study 49: 42–49. Threskiornis molucca at an urban breeding colony, south-east Queensland. Corella 26. Greenwood PJ, Harvey PH (1982) The natal and breeding dispersal of birds. 30: 41–45. Annual Review of Ecology and Systematic 13: 1–21. 12. Martin JM, French K, Major RE (2010) Population and breeding trends of an 27. Haas CA (1998) Effects of prior nesting success on site fidelity and breeding urban coloniser: the Australian white ibis. Wildlife Research 37: 230–239. dispersal: an experimental approach. Auk 115: 929–936. 13. Martin JM, French K, Major RE (2007) The pest status of Australian white ibis 28. Stracey CM, Robinson SK (2012) Are urban habitats ecological traps for a native (Threskiornis molucca) in urban situations and the effectiveness of egg–oil in songbird? Season-long productivity, apparent survival, and site fidelity in urban reproductive control. Wildlife Research 34: 319–324. and rural habitats. Journal of Avian Biology 43: 50–60. 14. Carrick R (1962) Breeding, movements and conservation of ibises (Threskior- 29. Kingsford RT (2000) Ecological impacts of dams, water diversions and river nithidae) in Australia. Wildlife Research 7: 71–88. management on floodplain wetlands in Australia. Austral Ecology 25: 109–127.

PLOS ONE | www.plosone.org 7 November 2012 | Volume 7 | Issue 11 | e50006 Behavioural Adaptation of an Urban Bird

30. Lemly AD, Kingsford RT, Thompson JR (2000) Irrigated agriculture and 46. Marzluff JM, McGowan KJ, Donnelly R, Knight RL (2001) Causes and wildlife conservation: conflict on a global scale. Environmental Management 25: consequences of expanding American crow populations. In: Marzluff JM, 485–512. Bowman R, Donnelly RE editors. Avian ecology and conservation in an 31. Australian Bird and Bat-banding Scheme (2009) Available: http://www. urbanizing world. Kluwer Academic, Norwell, MA. 331–363. environment.gov.au/cgi-bin/biodiversity/abbbs/abbbs-search.pl.Accessed 47. Evans KL, Newton J, Gaston KJ, Sharp SP, McGowan A, et al. (2012) 2009 Nov 04. Colonisation of urban environments is associated with reduced migratory 32. Porter JL, Kingsford RT, Hunter SJ (2006) Aerial survey of wetland birds in behaviour, facilitating divergence from ancestral populations. Oikos 121: 634– eastern Australia–October 2003–2005. NSW National Parks and Wildlife 640. Service Occasional Paper No.37. 48. Partecke J, Gwinner E (2007) Increased sedentariness in European blackbirds 33. Brandis KJ, Kingsford RT, Ren S, Ramp D (2011) Crisis water management following urbanization: a consequence of local adaptation? Ecology 88: 882–890 and ibis breeding at Narran Lakes in arid Australia. Environmental 49. Lank DB (1979) Dispersal and predation rates of wing-tagged semipalmated Management 48: 489–498. sandpipers Calidris pusilla and an evaluation of the technique. Wader Study 34. Marchant S, Higgins PJ (Eds.) (1990) Handbook of Australian, New Zealand and Group Bulletin 27: 41–46. Antarctic Birds. Vol. l: Ratites to –Part B. Australian Pelican to Ducks. 50. Coulson JC, Butterfield J, Duncan N, Thomas C (1987) Use of refuse tips by Melbourne: Oxford University Press. adult British herring gulls Larus argentatus during the week. Journal of Applied 35. Clergeau P, Yesou P (2006) Behavioural flexibility and numerous potential Ecology 24: 789–800. sources of introduction for the sacred ibis: causes of concern in western Europe? 51. Dingle H, Holyoak M (2001) The evolutionary ecology of movement. In: Roff Biological Invasions 8: 1381–1388. 36. Herring G, Gawlik DE (2008) Potential for successful population establishment DJ, Fairbairn RD, Fox C, editors. Evolutionary ecology: concepts and case of the non-indigenous sacred ibis in the Florida Everglades. Biological Invasions studies. New York, NY: Oxford University Press. 247–261. 10: 969–976. 52. Smith ACM (2009) Population ecology of the Australian white ibis, Threskiornis 37. Epstein JH, McKee J, Shaw P, Hicks V, Micalizzi G, et al. (2007) The Australian molucca, in the urban environment. PhD thesis. Sydney, Australia: University of white ibis (Threskiornis molucca) as a reservoir of zoonotic and livestock pathogens. Technology. EcoHealth 3: 290–298. 53. Kingsford RT, Roshier DA, Porter JL (2010) Australian waterbirds – time and 38. Martin JM, Major RE (2010) The use of cattle ear-tags as patagial markers for space travelers in dynamic desert landscapes. Marine and Freshwater Research large birds – a field assessment on adult and nestling Australian white ibis. 61: 875–884. Waterbirds 33: 264–268. 54. Tortosa FS, Caballero JM, Reyes-Lo´pez J (2002) Effect of rubbish dumps on 39. Corben D (2003) Population composition, abundances, and movements of the breeding success in the white in southern . Waterbirds 25: 39–43. Australian white ibis, Threskiornis molucca (), in an urban 55. Ridoutt C (2011) Organohalogenated pollutants in Australian white ibis environment. University of Technology Sydney, Honours Thesis. (Threskiornis molucca) eggs of urban through to inland sites. Honours thesis. 40. Pollock KH (1982) A capture–recapture design robust to unequal probability of Sydney, Australia: University of New South Wales. capture. Journal of Wildlife Management 46: 757–760. 56. Ji W, Sarre SD, White PCL, Clout MN (2004) Population recovery of common 41. White GC, Burnham KP (1999) Program MARK: survival estimation from brushtail possums after local depopulation. Wildlife Research 31: 543–550. populations of marked animals. Bird Study 46: 120–139. 57. Tuyttens FAM, Delahay RJ, MacDonald DW, Cheeseman CL, Long B, et al. 42. Burnham K, Anderson D (2002) Model Selection and Multi-Model Inference. Spring- (2000) Spatial perturbation caused by a badger (Meles meles) culling operation: Verlag. implications for the function of territoriality and the control of bovine 43. Pyke GH, Recher HF, O’Connor PJ (1989) Patterns of residency and movement tuberculosis (Mycobacterium bovis). Journal of Animal Ecology 69: 815–828. among honeyeaters in heathland near Sydney. Emu 89: 30–39. 58. Rodrigo KH, Bashford J, McCallum H, Jones M (2009) Contact networks in 44. Lowe KW (1984) The feeding and breeding biology of the sacred ibis Threskiornis a wild Tasmanian devil (Sarcophilus harrisii) population: using social network aethiopicus in southern Victoria. PhD thesis. Australia: University of Melbourne. analysis to reveal seasonal variability in social behaviour and its implications for 45. Belant JL (1997) Gulls in urban environments: landscape-level management to transmission of devil facial tumour disease. Ecology Letters 12: 1147–1157. reduce conflicts. Landscape and Urban Planning 38: 245–258.

PLOS ONE | www.plosone.org 8 November 2012 | Volume 7 | Issue 11 | e50006